The long-range objectives of this research are to investigate those procedures which will facilitate an automated analysis and classification of the cells in the peripheral blood. The proposed research is directed towards an automated analysis of the peripheral blood erythrocytes. Procedures will be developed for automatically classifying, clinically significant, individual abnormal erythrocytes and combining such classifications with other blood film information to obtain diagnostically relevant erythrocyte indices. The assumption is that the morphological characteristics of the individual cells will continue to be important diagnostic criteria in hematology, and that intuitively meaningful measurements of cellular morphology (shape, size, color, etc.) obtained by means of optical and digital image processing will be valuable inputs to the currently evolving automated clinical laboratories. BIBLIOGRAPHIC REFERENCES: Bacus, J. W.: A Whitening Transformation for Two-Color Blood Cell Images. Pattern Recognition. 8: 53-60, 1976. Bacus, J. W., Belanger, M. G., Aggarwal, R. K. and F. E. Trobaugh, Jr.: Image Processing for Automated Erythrocyte Classification. J. Histo. and Cyt. 24:(1), 195-201, 1976.